Artificial intelligence in retinal disease: clinical application, challenges, and future directions.

Retinal diseases are a leading cause of blindness in developed countries, accounting for the largest share of visually impaired children, working-age adults (inherited retinal disease), and elderly individuals (age-related macular degeneration). These conditions need specialised clinicians to interpret multimodal retinal imaging, with diagnosis and intervention potentially delayed. With an increasing and ageing population, this is becoming a global health priority. One solution is the development of artificial intelligence (AI) software to facilitate rapid data processing. Herein, we review research offering decision support for the diagnosis, classification, monitoring, and treatment of retinal disease using AI. We have prioritised diabetic retinopathy, age-related macular degeneration, inherited retinal disease, and retinopathy of prematurity. There is cautious optimism that these algorithms will be integrated into routine clinical practice to facilitate access to vision-saving treatments, improve efficiency of healthcare systems, and assist clinicians in processing the ever-increasing volume of multimodal data, thereby also liberating time for doctor-patient interaction and co-development of personalised management plans.

KCNV2-Associated Retinopathy: Detailed Retinal Phenotype and Structural Endpoints-KCNV2 Study Group Report 2.

PURPOSE: To describe the detailed retinal phenotype of KCNV2-associated retinopathy. STUDY DESIGN: Multicenter international retrospective case series. METHODS: Review of retinal imaging including fundus autofluorescence (FAF) and optical coherence tomography (OCT), including qualitative and quantitative analyses. RESULTS: Three distinct macular FAF features were identified: (1) centrally increased signal (n = 35, 41.7%), (2) decreased autofluorescence (n = 27, 31.1%), and (3) ring of increased signal (n = 37, 44.0%). Five distinct FAF groups were identified based on combinations of those features, with 23.5% of patients changing the FAF group over a mean (range) follow-up of 5.9 years (1.9-13.1 years). Qualitative assessment was performed by grading OCT into 5 grades: (1) continuous ellipsoid zone (EZ) (20.5%); (2) EZ disruption (26.1%); (3) EZ absence, without optical gap and with preserved retinal pigment epithelium complex (21.6%); (4) loss of EZ and a hyporeflective zone at the foveola (6.8%); and (5) outer retina and retinal pigment epithelium complex loss (25.0%). Eighty-six patients had scans available from both eyes, with 83 (96.5%) having the same grade in both eyes, and 36.1% changed OCT grade over a mean follow-up of 5.5 years. The annual rate of outer nuclear layer thickness change was similar for right and left eyes. CONCLUSIONS: KCNV2-associated retinopathy is a slowly progressive disease with early retinal changes, which are predominantly symmetric between eyes. The identification of a single OCT or FAF measurement as an endpoint to determine progression that applies to all patients may be challenging, although outer nuclear layer thickness is a potential biomarker. Findings suggest a potential window for intervention until 40 years of age.

Stargardt misdiagnosis: How ocular genetics helps.

Ocular Genetics at Wills Eye Hospital sees a wide range of rare disorders for accurate diagnosis. To demonstrate how focused consultation and genetic testing results in precise diagnoses, we investigated false diagnosis rates for patients referred with a diagnosis of Stargardt disease. This is a retrospective review of patients over a 3 year period referred to our Ocular Genetics clinic for possible Stargardt disease, or already holding a diagnosis of Stargardt disease. Results of diagnostic and genetic testing were compared to standard definition of Stargardt. Of 40 patients, 14 (35%) had been misdiagnosed. Four had non-Stargardt phenotype of which three had ABCA4 pathogenic variants with phenotypes inconsistent with Stargardt disease. Two of those with pathogenic ABCA4 variants were related. Nine had pathogenic variants in other different genes with overlapping features of Stargardt disease. One had Thioridazine maculopathy. Our study highlights the essential role of the subspecialty field of ocular genetics in obtaining accurate diagnoses for the delivery of correct counseling and interventional trial eligibility assessment.